CoreCMO

Measurement & Influence


Budget & Allocation

Defending the budget. Line by line, channel by channel, with the math. The artifact you bring to the CFO before the budget conversation, not during.

Measurement & Influence 3 prompts 1 agent — Budget Allocation Agent ~5 min preview

The framework — strategy first


Budget & Allocation — the strategic foundation.

STRATEGY & PROCESS

Marketing budget planning bridges strategy and execution. This work defines how you allocate, manage, and report on marketing investment to maximize return and maintain clear accountability.

The 80/20 brand-vs-demand rule (the sales-led version).

The single most-debated allocation question in B2B SaaS marketing for any Chief Marketing Officer (CMO): how much budget goes to brand vs demand? The framework that works at scale for sales-led companies: allocate 20% of program and headcount to brand, 80% to demand. Then focus on using the 80% of demand budget to drive 100% of demand outcomes. Brand is the big bet.

This isn't about under-investing in brand — it's about being honest about what brand can move on a 90-day timeline (almost nothing measurable) vs what demand can move (every quarterly Key Performance Indicator, or KPI, you'll be measured on). Both work together. Brand makes demand cheaper over time. Demand pays the rent.

THE CONTINGENCY — 80/20 BENDS BY MOTION

The 80/20 rule is the sales-led default. It bends for PLG and for community-led.

PLG → 10/90. The product is the brand expression. Classic brand campaigns (LinkedIn rebrands, billboards, the annual film) underperform in PLG because the buyer's "brand experience" is the first 90 seconds of the app. Brand spend goes into developer relations, category education, and design-system polish — not media buys. The 10% to brand at a Notion or Linear is mostly design talent + documentation craft.

Community-led → 30/70. Community-led companies invest more in brand because community-building IS partly brand work. The conference, the magazine, the awards program, the sponsorships — all show up on the brand line, all drive the actual demand. Treat the 30% as a higher floor than the standard, not as inefficiency.

Vertical SaaS → carve out a "vertical authority" line. Horizontal 80/20 framing under-funds vertical authority (industry-specific research, vertical events, named vertical advisory boards) which is what wins concentrated mid-market verticals. Hold the 80/20 envelope but ring-fence 15–20% of the brand line for vertical-authority bets.

The 80/20 rule, applied

  • Headcount: Allocate roles by % brand vs % demand. A designer might be 50/50. A demand-gen manager is 100% demand. A content marketing lead might be 30/70. Capture the split in your budget tracker so you can defend the brand investment line item if it's challenged.
  • Programs: 80% of programs budget funds demand engines (paid, content, lifecycle, ABM, events). 20% funds brand bets — original research, category-defining campaigns, executive thought leadership, design investments.
  • The rule for brand investment: Brand isn't where you spread peanut butter — it's where you make 1–3 big, defensible bets per year. The annual benchmark report. The category-defining content series. The executive-led thought leadership program. Not 27 small things.
  • The rule for demand: Use the 80% to drive 100% of demand outcomes. If you can't drive your number with 80% of budget on demand, the channel mix is wrong, not the budget envelope.

Allocation by segment — the budget conversation that earns trust with finance

The Chief Financial Officer (CFO)'s first question is "where's the money going?" — answer it before they ask. Build an allocation-by-segment table that shows program spend, payroll allocation, total marketing spend, allocation %, and the resulting pipeline and bookings outcomes per segment. The math tells the story.

SEGMENTPROGRAMS & DEMANDPAYROLL & ALLOCATIONSTOTAL SPENDALLOCATION %PIPELINE GENERATEDBOOKINGS$1 INVESTED → $X PIPELINE$1 INVESTED → $X BOOKINGS
Segment 1 (e.g., Enterprise)$1,234,898$1,234,898$2,469,79658%$26,405,933$14,758,969$10.69$5.98
Segment 2 (e.g., Mid-market)$890,000$890,000$1,780,00042%$10,502,030$2,402,933$5.90$1.35

The table above is a worked example. The pattern: a segment is healthy when $1 invested generates $5+ in pipeline and $1.50+ in bookings. Anything below those ratios means you've over-allocated to that segment or under-monetized the channel mix. Quarterly review with the CFO. Reallocate based on what the numbers say, not what the loudest team member says.

Allocation by function — the canonical nine-line view your CFO will accept

Segment allocation tells the revenue story. Function allocation tells the operations story. Show both. Most B2B SaaS marketing budgets sort cleanly into nine functions: Events & Field, Demand Gen, Content & SEO, Brand & Creative, PR & Analyst Relations, Product Marketing (incl. win/loss + analyst research), Customer Marketing, Marketing Ops, and Tools & Subscriptions. Allocating to all nine is what separates a defensible budget from one that gets death-by-a-thousand-cuts.

Benchmark sources for the % ranges below: OpenView SaaS Marketing Benchmarks (2024), KeyBank Software Industry Report (2024), Pavilion CMO Allocation Survey (2024–2025), Spencer Stuart CMO Compensation & Spend Tracker, CMI B2B Content Marketing Trends (2024). Ranges represent the inter-quartile range observed across $10M–$100M ARR B2B SaaS; your row will vary by motion, vertical, and stage. Treat as planning anchors, not exact targets.

FUNCTIONTYPICAL % OF TOTALWHAT IT FUNDS
Events & Field15–25%Tier 1 flagship, sponsored conferences, owned events, field activations, booth + travel + logistics. Often the largest single line at sales-led B2B SaaS companies.
Demand Gen20–30%Paid media (Google, LinkedIn, Meta, 6sense), content syndication, webinars, ABM platform, review-site programs. The "pipeline now" function.
Content & SEO8–12%Editorial team, freelance network, SEO tooling, content production. Compounds over time — one of the highest long-term ROI lines.
Brand & Creative5–10%Brand book, design, video, agency-creative retainers, photography, big bets. Lower at Series A, higher post-Series C.
PR & Analyst Relations6–12%PR agency or in-house lead, press releases / newswire, analyst subscriptions (Gartner / Forrester / IDC / vertical), industry awards, executive thought leadership, original research reports. The line that compounds longest and gets cut first.
Product Marketing4–8%Win/loss research, custom analyst projects, competitive intel tooling, sales enablement asset production. Often partly funded under Product budget instead.
Customer Marketing3–8%Advocacy program, case-study production, customer community + events, referral programs, expansion campaigns.
Marketing Ops2–5%Agency retainer for web maintenance + automation, integration work, attribution tooling configuration.
Tools & Subscriptions3–6%The full stack: MAR, CRM contributions, intent data, SEO tools, design tools, social tools, BDR enrichment, analytics. Compounds quickly — audit annually.

Ranges shift by motion: PLG skews lower on Events and higher on Content + Product Marketing; ABM skews higher on Events and Demand Gen; sales-led sits mid-range across the board. Read your row, then apply the motion adjustment.

PR & AR — THE LINE CFOs CUT FIRST AND CMOs REGRET CUTTING MOST

The 6–12% PR/AR allocation is non-negotiable for any company at $10M+ ARR that sells to enterprise. Below that band, PR can be lighter; above it, defunding PR/AR is among the most expensive savings in a CMO's career. Reference points from real budgets:

Dollar examples are operator anchors from 2024–2026 PE-backed portfolios; your retainer ranges shift by firm tier and seasonal demand. Treat as starting point for the conversation, not a benchmark.

  • Series B vertical SaaS at ~$2M total marketing spend. Typical mix: PR agency retainer in the $60–80K/yr band (3–4%), analyst & research in the $70–90K/yr band (3.5–4.5%), an industry awards program at $20–30K (1–1.5%), an annual proprietary research report at $40–60K (2–3%), totaling roughly 10–12% of budget. That’s the upper-middle of the range and the right calibration when the company is competing for enterprise Requests for Proposal (RFPs) and analyst inclusion matters.
  • Series A horizontal SaaS at ~$1M total. Drop the Gartner subscription, keep a smaller PR retainer ($3–4K/mo), keep the awards program, run analyst inquiry calls but skip the full briefing rhythm. Target: 6–8%.
  • Series C+ enterprise SaaS at $5M+ total. Full AR program (Gartner, Forrester, IDC, vertical analyst), in-house PR lead, retained agency for amplification, original research as a flagship annual bet. Target: 10–12%.

The function deserves its own line — not scattered under Brand and Product where it gets invisible during a 15% cut conversation. Cite the AI-era PR framing in PR & Comms when the CFO asks why this can't be a Q4 deferral: this is the channel the buyer's AI tools consult, and the analyst report cited in next year's enterprise RFPs. Cut it and pay for it twice when sales misses Q3 next year.

The agency rule — supplemental, not core

Agencies are the most over-used line item in early-stage marketing budgets. Use them right and they accelerate; use them wrong and they're a substitute for the team you should have hired. The rule: use agencies as supplemental support when you're scaling. You don't need to hire EVERYONE in-house, but the strategic IP — brand voice, ICP, positioning, agent specs — has to live in-house. Agencies are great for execution at scale; they're terrible owners of strategy.

The acceptable agency uses: paid media execution and optimization, design at scale, video production, PR/AR (until you have an in-house lead), event production. The not-acceptable: anything that requires the agency to learn your brand voice from scratch every quarter, anything where the strategic decision belongs to you, anything where the agency owns the relationship with the buyer.

The downloadable budget template — 24 sheets, formulas, benchmarks.

Pick the template that matches your stage. Each one carries the same 24-sheet structure (so your mental model carries across stage transitions) with stage-appropriate defaults for function allocation, headcount, brand-bet count, and tooling. Sheet 24 cites every benchmark source (OpenView, KeyBank, Pavilion, Spencer Stuart, CMI, G2) so your numbers are CFO-defensible.

Series A · $1–3M

corecmo-budget-series-a.xlsx

Demand-heavy, 1 brand bet, agency-supplemented. ~30% of revenue. 4 default FTEs. ~$40K tooling.

Series B–C · $3–15M · Most common

corecmo-budget-series-bc.xlsx

Building proper teams. PR/AR begins. ABM tooling enters. 3 brand bets. ~18% of revenue. 10 default FTEs.

Series D+ / PE-backed · $15–50M

corecmo-budget-series-d-pe.xlsx

Full team. Strong AR program. Mature attribution. 4 brand bets. ~14% of revenue. 14 default FTEs.

Public · $50M+

corecmo-budget-public.xlsx

Investor-optics layer. Multi-product allocation. Quarterly forecast tied to Wall Street. 5 brand bets. ~11% of revenue.

WHAT’S INSIDE THE 24 SHEETS

Summary · Function Allocation (the canonical 9-line) · Headcount & Comp · Quarterly Phasing · Monthly Cash Flow · Variance Tracker · Per-function detail (9 sheets) · Tech Stack · Agency Roster · Brand Bets · Pipeline Math by Channel · ROI by Channel · Scenario Modeling (Plan / Down 10% / Down 20%) · Board Snapshot (auto-pulled) · CFO Defense (anticipated objections + prepared responses) · Benchmarks Reference (source-cited).

Formulas wired throughout. Update one cell on Sheet 1 (your revenue) and the entire budget flows. Update Function Allocation on Sheet 2 and the per-function detail sheets, quarterly phasing, scenario model, and board snapshot all recalculate.

Your Budget Allocation

Your real numbers. Saves to your Brief — every budget, paid media, and CFO-conversation prompt will use these.

Saved as [ANNUAL MARKETING BUDGET], [BRAND ALLOCATION], [DEMAND ALLOCATION], [BRAND BETS], [AGENCY ROSTER].

The Brand Bet rubric — defending (or vetoing) the 20%

The 20% brand line is the most-requested and least-defended budget. CFOs cut it first because the math is hardest to defend. Use the rubric below to defend it, or to veto your own request, before it lands on the board deck. Run every brand-bet candidate (billboard, sponsored research, podcast, exec-LinkedIn program, conference takeover) through the same gates.

The 4-fit check — is this brand bet for THIS company?

  1. Audience fit. The buyers you want to reach are either massive enough that broad-reach mediums hit them or tightly concentrated enough that geo-targeted ones do. Scattered B2B audiences turn brand-bet dollars into waste.
  2. Business model fit. Your ACV and motion can pay back blunt awareness. PLG works if the audience is broad and the funnel converts. Sales-led works if a single influenced deal makes the spend back. Pre-PMF almost never works.
  3. Execution fit. Your brand and creative are sharp enough that the bet will turn heads, not blend in. If brand voice is still being coalition-built, no agency fixes that for you.
  4. Budget fit. You can afford enough of the bet to be seen (not one rotation on a digital unit, not one sponsored research piece nobody saw) AND cover production AND not rob demand. Opportunity cost is the line most marketers skip. Ask: what else could this dollar do?

Miss any one, and the bet is a flex, not a brand investment.

The 4 honest goals — name the one

Every brand bet should map to one goal. Force the requesting executive to pick.

  • "We've arrived." Build awareness for a brand the market doesn't yet recognize.
  • "We're still here." Maintain awareness; often paired with a launch or event.
  • "We've made it, insiders." Celebrate with the team or marquee customers (social moment, employee photo, customer feature).
  • "Come join us." Name recognition with candidates, investors, or analysts in a specific market.

If the real reason is "we just want to," name it. Write down what success looks like. Cap the spend so it doesn't eat the demand engine. That's a legitimate move once or twice in a CMO tenure. It's never a quarterly habit.

Sticker shock — what brand bets actually cost in 2026

  • Bay Area freeway billboard: $20K–50K/month per board. A 3-month campaign with creative production is $80K–200K. At 14% of revenue, that's the entire annual brand line for a $5M ARR company.
  • NYC subway station takeover: $40K–120K for a 4-week campaign.
  • Original benchmark research report: $40K–150K including survey panel, analyst time, design, and the launch motion that makes it land. Compounds for 12+ months if syndicated and SEO'd.
  • Category-defining podcast: $80K–250K/year for a credible 26-episode season with guest production, editing, and amplification.
  • Hand-painted wall (SF, NYC, LA): $30K–90K, depending on location, size, and permit complexity. Memorable but rarely measurable.

Cross-reference the Out-of-Home callout in Paid Media before approving any OOH-shaped request. The math doesn't live in this chapter, but the discipline does.

Budget Philosophy

CORE PRINCIPLES

  • Invest against pipeline, not activities. Budget is allocated to programs with demonstrated or projected pipeline contribution, not to maintain channel presence.
  • The 70-20-10 rule. 70% of budget funds proven, high-ROI channels; 20% tests emerging channels; 10% explores experimental bets.
  • Monthly reforecasting. Budget is not set-and-forget. Monthly variance reviews drive reallocation toward what's performing.
  • Full-funnel allocation. Budget spans awareness (10–20%), consideration (30–40%), and decision (40–50%) stages — weighted by current funnel gaps.
  • Attribution transparency. Every dollar is tracked to a program; every program is measured against pipeline or MQL contribution.

Industry Benchmarks: Marketing as % of Revenue

Benchmarks below are operator observations from working across PE-backed B2B SaaS; current OpenView/Pavilion/KeyBank surveys are directionally similar. Treat as planning anchors, not exact targets.

STAGE / ARR BANDMARKETING AS % OF REVENUEMARKETING % OF TOTAL OPEXNOTES
Seed / Pre-PMF25–40%20–30%Heavy brand + product-market fit experiments
Series A ($1M–$10M ARR)20–35%20–25%Build demand gen engine; paid + content investment
Series B ($10M–$30M ARR)15–25%15–20%Scale proven channels; hire specialized team
Series C ($30M–$75M ARR)12–20%12–18%Efficiency pressure begins; track CAC payback
Growth ($75M–$150M ARR)10–15%10–15%Efficiency > growth; optimize CAC:LTV ratio
Scale ($150M+ ARR)8–12%8–12%Category leadership; brand + enterprise sales motion

Allocation KPIs by function — what each line should be measured on

The percentages above tell you where the money goes. The table below tells you how each line proves it worked. Same nine functions as the canonical allocation table — this is the measurement layer.

MARKETING FUNCTIONPRIMARY KPISNOTES
Events & FieldAttendees, pipeline from events, 10× event ROIInclude sponsorship, production, and lead follow-up costs
Demand GenMQLs, pipeline sourced, CPLOptimize via the channel mix model in Paid Media
Content & SEOOrganic traffic, content MQLs, keyword rankingsCompounds over time — protect the line through quarterly squeezes
Brand & CreativeShare of Voice, brand recall, direct web trafficLonger payback; tied to the 1–3 brand bets you've named for the year
PR & Analyst RelationsTier 1 placements, analyst mentions, SOVPR agency retainer + AR program; the line that compounds longest
Product MarketingWin rate, competitive displacement, sales velocityPrimarily headcount + agency; under-invested at most companies
Customer MarketingNRR, expansion pipeline, customer stories producedExpansion revenue is lowest-cost ARR — under-fund at your peril
Marketing OperationsAttribution accuracy, tech stack ROI, campaign velocityMarTech stack + RevOps headcount allocation
Tools & SubscriptionsTool ROI, redundancy elimination, contract optimizationAudit annually — compounds quickly without discipline

Annual Budget Planning Process

Bottom-Up Demand Plan (Week 1–2): Start with the revenue target. Work backwards from pipeline coverage ratio to determine required MQL and pipeline volumes by channel.

Channel-Level ROI Analysis (Week 2–3): Pull prior year performance by channel. Calculate CPL, cost-per-pipeline, and cost-per-acquisition per channel. Kill underperformers. Double down on proven channels.

Headcount Plan (Week 2–3): Align marketing headcount plan with Finance's approved headcount. Loaded cost per hire is included in total marketing budget.

Program Budget Build (Week 3–4): Build campaign-level budget plans. Each campaign should have: estimated spend, projected MQLs, projected pipeline, ROI assumption.

Executive Review & Alignment (Week 4–5): Present to CEO/CFO. Tie every budget line to a business outcome. Have an alternative 'downside scenario' showing 15% cut impact.

Board Presentation (Week 5–6): Present marketing plan and budget at Board or Audit Committee as applicable. Include benchmark data vs. industry and prior year.

Lock & Publish (Week 6): Finalize budget. Publish to marketing team. Create monthly tracking framework in finance system.

Monthly Budget Review Cadence

REVIEW TYPECADENCEPARTICIPANTSKEY QUESTIONS
Budget vs. ActualsMonthly (Day 10–15 of following month)Marketing leads + FinanceWhere are we over/under? What reallocation is needed?
Channel Performance ReviewMonthlyDemand Gen + RevOpsWhich channels are performing vs. target? CPA trending?
Pipeline Attribution ReportMonthlyCMO + CRO + RevOpsIs marketing hitting pipeline targets? What's the quality?
Quarterly Budget ReforecastQuarterlyCMO + CFOReallocate toward outperforming channels; reduce underperformers
Annual Planning KickoffQ3 (for following year)CMO + Finance + CEOBegin top-down revenue targets + bottom-up marketing plan

Campaign-Level Budget Template

CAMPAIGN BUDGET SCORECARDCAMPAIGN NAME:CAMPAIGN GOAL (AWARENESS / MQL / PIPELINE):TARGET AUDIENCE:CAMPAIGN DATES:TOTAL BUDGET REQUESTED: $BREAKDOWN — CONTENT/CREATIVE: $ \PAID DISTRIBUTION: $ \EVENTS/PRODUCTION: $ \AGENCY/FREELANCE: $PROJECTED MQLS:PROJECTED PIPELINE (AT AVG ACV × WIN RATE):PROJECTED COST-PER-MQL: $PROJECTED ROI (PIPELINE / TOTAL SPEND): ×BUDGET OWNER:FINANCE PO/COST CENTER:

Budget Defense & Benchmarking

When CFOs push back on marketing spend, Marketing must defend investment with data — not intuition.

BUDGET DEFENSE TOOLKIT

  • Pipeline-coverage proof. Show that every $1 of marketing spend generates $X of pipeline. If pipeline sourced + influenced exceeds budget by 5–10×, marketing is generating positive leverage.
  • CAC payback analysis. Show CAC payback period vs. industry benchmark. If you're at 12 months vs. an 18-month industry average, marketing is efficient.
  • Channel-by-channel ROI. Show cost-per-pipeline by channel. Defend the top 3 channels with hard ROI data. Offer to cut the bottom 2.
  • Competitor spend benchmark. Use publicly available data and analyst reports to show what peers are spending as a % of revenue.
  • Revenue attribution model. Show a clear, audited attribution model that ties marketing programs to closed-won revenue — not just MQLs.

Marketing Budget KPIs

KPIFORMULATARGET
Marketing Efficiency Ratio (MER)Pipeline Sourced / Total Marketing Spend> 5×
Cost Per Lead (CPL)Total Spend / Total MQLsTrack by channel; total CPL < $[BENCHMARK]
Cost Per Qualified OpportunityTotal Spend / SALs or SQLs created< 3× CPL; improving QoQ
Cost Per Acquisition (CPA)Total Marketing Spend / New Customers from Marketing Sources< 30% of LTV
Budget Utilization RateActual Spend / Budgeted Spend90–105% (under-spend = opportunity cost; over-spend = control issue)
ROI by CampaignPipeline Generated / Campaign Spend> 10× ROI for demand gen campaigns

The prompt pack


Paste-ready prompts for Budget & Allocation.

Each prompt is a named, named-by-what-it-does deliverable. Click any card to expand the paste-able body. Run against your Operator Brief.

Five copy-paste prompts. Open ChatGPT, Claude, or Gemini. Paste a prompt. Run it. The output of one prompt feeds into the next.

READ THIS ONCE BEFORE ANY PROMPT IN THIS BOOK

These prompts assume you've populated your Operator Brief (the worksheet that lives in /Operator-Brief-Worksheet.docx). When a prompt asks for OPERATOR BRIEF, paste the relevant Brief sections rather than typing context from scratch.

Your output then arrives in your voice, against your buyers, using your differentiators. Not [BRACKETED] generics. The Brief is the difference between an LLM helper and a tool that sounds like you.

Prompt 1

Allocation by stage

A budget allocation across funnel stages with rationale and benchmarks.

Allocate our marketing budget across funnel stages. TOTAL BUDGET: $[] STAGE: [] ARR: $[] CURRENT MOTION: [] Allocate across Awareness / Demand / Pipeline / Customer / Brand / Ops. % per bucket + dollar amount. Cite ARR-band benchmark. Defend the largest line in 2 sentences. Flag the line that's underfunded.

Prompt 2

Campaign budget scorecard

A scorecard for evaluating new campaign budget requests.

Score this campaign budget request. CAMPAIGN: [] ASK: $[] OWNER: [] HYPOTHESIS: [] KILL CRITERIA: [] Score 1-5 on: hypothesis clarity, kill criteria, audience fit, expected pipeline, opportunity cost. Recommend approve / reduce / decline + rationale (1 sentence).

Prompt 3

Budget defense memo

A 1-page memo defending the marketing budget to the CFO.

Defend our marketing budget for next quarter. CURRENT BUDGET: $[] REQUESTED: $[] RECENT PIPELINE INFLUENCE: [] ARR: $[] 1-page memo to CFO. Sections: pipeline created vs spend (the ratio), wins from last quarter (specific deals), the ROI of the marginal $50k, what we'd cut first if forced. Tone: peer-to-peer, not pleading.

The agent


Budget & AllocationAgent.

For teams operating in AI Operating Model territory, the work above runs on a named agent that watches the inputs on cadence and surfaces drift before the human team notices.

How to install this agent

Five steps from spec to running agent.

  1. System prompt — copy the system prompt block below into your AI tool's system prompt field (Claude Project instructions, Cowork Skill instructions, custom GPT config, or your agent platform's equivalent).
  2. Inputs — wire the inputs as the agent's reference files. The Operator Brief is always input #1; the other inputs vary by agent.
  3. Outputs — the output schema tells you what the agent produces. Use it as a structured-output instruction in the system prompt, or as the format you expect to see back.
  4. Evals — before publishing any output, score it against the eval criteria. Don't ship anything that doesn't pass.
  5. Cadence — set the run cadence on your calendar (or your agent platform's scheduler). Log every run in your wins log.

Budget Allocation Agent

Watches budget pacing across every line in real time. Surfaces over- and under-pacing weekly. Validates new spend asks against pipeline math + saturation curves. Defends the budget envelope from the cumulative drag of small ‘just this once’ exceptions.

Who is this agent
Identity card
NameBudget Allocation Agent
RoleBudget pacing + spend governance — the envelope-protection layer
OwnerDirector of Marketing Operations (with CFO co-ownership)
Reports toVP Marketing + CFO
Versionv0.5 (supervised)
SurfaceReplit + Postgres (budget envelope + actuals ledger) + integration with finance system + Performance Marketing Agent
Output target/budget/weekly-pacing.md + /budget/spend-asks-queue.md + /budget/quarterly-reconciliation.md
Review cadenceDaily pacing snapshot; weekly variance review; monthly forecast reconciliation; quarterly budget re-plan
Mission
Defend the budget envelope without becoming the budget bureaucracy. Watch pacing daily across every line. Surface over- and under-pacing weekly. Validate new spend asks against pipeline math + MMM saturation curves — would this dollar actually do work? Be the agent that catches small ‘just this once’ exceptions before they accumulate into a quarter-end overrun.
Goals & KPIs the agent moves
Leading indicators — the agent controls these
Spend-ask routing within SLA. Up to 0.5% of monthly budget envelope auto-approves; above pages CFO with full reconciliation context< 48 hours auto-approve, < 5 business days CFO-gated
Reconciliation gap (agent actuals vs. CFO ledger) at month close< 1%
Lagging indicators — downstream outcomes with review triggers
Quarterly spend variance vs. plan. Trigger: 2 consecutive quarters outside ±7% pages the CFO + VP Marketing for budget-discipline review.Within ±5%
Over-pacing line items caught + corrected within 14 days of breach. Trigger: any quarter below 90% catch-rate pages the Director MarOps for pacing-controls review.≥ 95% caught + corrected
What it does
Task list
  1. Daily Pull yesterday’s spend across every line. Reconcile against the budget envelope. Compute trailing pacing per line.
  2. Daily Cross-check Performance Marketing Agent platform spend vs. CFO ledger. Flag any gap > 2%.
  3. Daily Watch new spend asks queue. Validate against pipeline math + MMM saturation curves. Route to approver.
  4. Weekly Compile weekly Budget Pacing digest — over-pacing lines, under-pacing lines, top concerns, recommended corrections.
  5. Weekly Forecast end-of-quarter spend based on current pacing. Surface gap to plan.
  6. Monthly Reconciliation against finance system. Walk every line. Surface any gap for resolution.
  7. Monthly Vendor + agency review: which spend lines are returning < 1× estimated value? Surface for renegotiation or cutoff.
  8. Quarterly Pre-quarter budget re-plan: based on pipeline math + saturation curves + AOS gap remediation, draft the next quarter envelope.
  9. Event When Performance Marketing Agent proposes a reallocation, validate against pacing + envelope before approving.
  10. Event When Best-in-Class Assessment Agent surfaces a remediation plan with required budget, queue for next budget cycle.
  11. Event When a spend line breaches its monthly cap, hard-hold further spend until Director MarOps + CFO approve.
Schedule grid
TaskFrequencyDurationOutput goes to
Daily pacing snapshotDaily 06:00~20 minDirector MarOps + CFO if variance
Daily Performance Marketing cross-checkDaily 07:00~15 minDirector MarOps + Performance Marketing Agent
Daily spend asks queue reviewDaily 16:00~15 minDirector MarOps + ask submitters
Weekly Budget Pacing digestWeekly Mon 09:00~30 minVP Marketing + CFO
Weekly end-of-quarter forecastWeekly Mon 09:30~30 minCFO + VP Marketing
Monthly reconciliationMonthly 5th~2 hoursDirector MarOps + CFO
Monthly vendor + agency reviewMonthly 15th~90 minDirector MarOps + VP Marketing
Quarterly pre-quarter re-planQuarterly Q-5 days~6 hoursVP Marketing + CFO + CEO
Triggers

Scheduled (cron-style):

ScheduleWhat it runs
0 6 * * *Daily pacing snapshot
0 7 * * *Daily Performance Marketing cross-check
0 16 * * *Daily spend asks queue review
0 9 * * 1Weekly digest + forecast
0 9 5 * *Monthly reconciliation
0 9 15 * *Monthly vendor review

Event-driven:

EventWhat it runs
Spend line breaches monthly capHard-hold further spend on that line; page Director MarOps + line owner
Daily reconciliation gap > 2%Open ticket + page Director MarOps within 4 hours
Performance Marketing Agent proposes reallocation > $5KValidate against pacing + envelope before signing off
End-of-quarter spend forecast variance > 10% from planPage VP Marketing + CFO; trigger mid-quarter correction options
New spend ask submittedValidate against pipeline math + saturation; route to approver within 24 hours
Who it works with
Inputs
SourceTypeCadenceRequired?
Operator Brief (Sections 1, 7)MarkdownRead on quarterly re-planRequired
Budget envelope (per-line annual + per-quarter)YAML / spreadsheetAnnual + quarterlyRequired — core config
Finance system ledger (NetSuite / QuickBooks / SAP)APIDailyRequired
Performance Marketing Agent platform spendAPIDailyRequired
Pipeline Math Agent forecast dataJSONWeeklyRequired for spend-vs-pipeline math
MMM Agent saturation curvesJSONMonthlyRequired for spend ask validation
Best-in-Class Assessment Agent remediation plansMarkdownQuarterlyRequired for budget-cycle planning
Vendor + agency contract registryPostgresContinuousRequired
Outputs
OutputFormatTarget pathAudience
Daily pacing snapshotJSON + Markdown/budget/daily-pacing.jsonDirector MarOps + CFO (Slack)
Weekly Budget Pacing digestMarkdown + chart/budget/digests/YYYY-WW.mdVP Marketing + CFO + line owners
Weekly end-of-quarter forecastMarkdown/budget/forecast/YYYY-WW.mdCFO + VP Marketing
Spend ask decision logMarkdown + approval trail/budget/spend-asks/YYYY-MM.mdDirector MarOps (audit) + ask submitters
Monthly reconciliation reportMarkdown/budget/reconciliation/YYYY-MM.mdDirector MarOps + CFO
Quarterly pre-quarter re-planMarkdown + spreadsheet/budget/quarterly/Q<n>-plan.mdVP Marketing + CFO + CEO + Board if material
↑ Upstream — agents/sources that feed this one
  • Operator Brief (human-maintained). KPI targets anchor the spend-vs-outcome justifications.
  • Pipeline Math Agent. Pipeline gap drives the spend-justification math.
  • MMM Agent. Saturation curves drive the ‘does the next dollar do work’ check.
  • Performance Marketing Agent. Platform spend ground truth.
  • Best-in-Class Assessment Agent. Remediation plans with required budgets for next-quarter planning.
  • Signal Router. Routes spend ask submissions + finance ledger events.
↓ Downstream — agents/humans that consume its output
  • Director MarOps + CFO (humans). Approve spend asks; sign off on quarterly re-plan.
  • Performance Marketing Agent. Receives reallocation validation + envelope context.
  • Line owners (humans). Receive over-pacing alerts on their lines.
  • VP Marketing + CFO. Weekly + monthly + quarterly reporting.
  • Best-in-Class Assessment Agent. Receives quarterly spend data for AOS budget-quality dimension.
Human escalation paths
Trigger conditionEscalate toWithin
End-of-quarter forecast variance > 10% from planVP Marketing + CFOSame week
Spend line breaches monthly capDirector MarOps + line owner + CFOSame business day
Daily reconciliation gap > 5% sustained 3+ daysDirector MarOps + CFO< 4 hours
Spend ask > $10K with insufficient pipeline-math justificationVP Marketing + CFO< 5 days
Vendor + agency review surfaces line returning < 1× valueVP Marketing + Director MarOpsSame month (renegotiation or cutoff decision)
How to build it
System prompt
You are the Budget Allocation Agent for [COMPANY]. YOUR JOB Defend the budget envelope without becoming bureaucracy. Watch pacing daily. Validate spend asks against pipeline math + saturation curves. Catch small "just this once" exceptions before they accumulate into overruns. INPUTS (always read in this order) 1. /operator-brief.md (Sections 1, 7) 2. /budget/envelope.yaml - per-line annual + per-quarter budget 3. /finance/ledger.json - finance system daily snapshot 4. /paid/spend-actuals.json (Performance Marketing Agent) 5. /pipeline-math/daily-snapshot.json (Pipeline gap context) 6. /mmm/saturation-curves.md (Spend ask validation) 7. Spend ask payload (when validating) OUTPUTS - /budget/daily-pacing.json (daily) - /budget/digests/YYYY-WW.md (weekly) - /budget/spend-asks/YYYY-MM.md (decision log) - /budget/reconciliation/YYYY-MM.md (monthly) - /budget/quarterly/Q<n>-plan.md (quarterly re-plan) SPEND ASK DECISION RULES - Ask <= $10K: validate against monthly pacing; auto-approve if line < 90% monthly cap and no over-pacing trend. - Ask $10K-$25K: validate against pipeline math (does this fill a known gap?) + saturation curve (does the next dollar do work?). Director MarOps approval. - Ask > $25K: same validation + CFO approval. - Any ask that breaches monthly cap: hard-hold; require CFO sign-off path. RULES 1. Every decision cites: pacing state, pipeline-gap context, saturation reading. 2. Never auto-approve outside envelope. Hard rule. 3. Reconciliation gap > 2% sustained 3 days = ticket. 4. Forecast variance > 10% from plan = page VPM + CFO. ESCALATION - Monthly cap breach: page Director + line owner + CFO same business day. - Reconciliation gap >5% sustained 3d: page Director + CFO <4h. - Forecast variance >10%: page VPM + CFO same week.
Tools & integrations
Platform / toolUsed forRequired?
Replit + Postgres (budget envelope + actuals ledger)State + audit trailRequired
Finance system API (NetSuite / QuickBooks / SAP / Sage Intacct)Daily ledger snapshotRequired
Performance Marketing Agent platform spendCross-checkRequired
Pipeline Math Agent APIPipeline gap contextRequired
MMM Agent APISaturation curve contextRequired
Slack APIDaily alerts + weekly digestRequired
Spreadsheet (Google Sheets / Excel)CFO-facing quarterly planRequired
Guardrails — what it must not do
  • Never auto-approve outside envelope. Hard rule.
  • Never modify the envelope autonomously. Envelope changes are quarterly + CFO-approved.
  • Never share per-line spend outside the marketing + finance scope.
  • Never use spend data to score individual line owners for performance management without VP Marketing sign-off.
  • Honor purchase-approval thresholds — ask values matter, not just total spend.
  • Never validate a spend ask without showing the pipeline-gap math + saturation reading.
  • Never round down variance. If it’s > 5%, it’s > 5%.
Evals + hallucination defense

Evals — output quality checks:

  1. Quarterly variance. Quarterly: actual spend variance vs. plan. Target within ±5%.
  2. Reconciliation accuracy. Monthly:Agent actuals vs. CFO ledger. Target < 1% gap.
  3. Approval latency. Weekly: p99 latency by ask-size band. Target < 48h for ≤ $10K; < 5 days for > $10K.
  4. Over-pacing catch. Quarterly: % of monthly-cap breaches caught + corrected within 14 days. Target 100%.

Hallucination defense — specific checkpoints:

  • Pacing claims must trace to specific ledger line items + transaction dates.
  • Spend ask validations must cite the pipeline math + saturation curve sources.
  • Reconciliation gaps must cite the specific lines + amounts.
  • Vendor + agency review must cite contract terms + outcome measurements.
  • When data is missing for a line, surface the gap rather than estimate.
Maturity curve + first-run checklist
v0.1 — Manual-assistDirector MarOps runs pacing by hand with agent assistance. Useful from day 1 for replacing spreadsheet budget tracking.
v0.5 — SupervisedDaily pacing + weekly digest + monthly reconciliation autonomous. Director MarOps + CFO sign off on quarterly re-plan. Default ship state.
v1.0 — Semi-autonomousAfter 4 quarters clean evals + variance consistently < 5%, agent can auto-approve spend asks ≤ $5K without Director MarOps gate. Asks > $5K always require human approval.

First-run checklist — 5 steps from spec to running agent:

  1. Author the budget envelope YAML with VP Marketing + CFO. Per-line annual + per-quarter caps.
  2. Wire finance system + Performance Marketing Agent + Pipeline Math + MMM inputs.
  3. Run reconciliation in shadow mode for 30 days. Director MarOps validates against finance team’s reading.
  4. Turn on live. Subscribe Director MarOps + CFO + VP Marketing to weekly digest. Spend asks queue lives in Slack #budget-asks.
  5. Schedule the quarterly pre-quarter re-plan cadence. Log every run in /budget/agent-log.md.
Strategic · Creative · Data Driven · Revenue Accelerator